2020 Case Studies in Data Analysis Competition
The Case Studies in Data Analysis Poster Competition will be held during the Annual Meeting of Statistical Society of Canada (SSC) at Carleton University. The case studies are intended to provide enthusiastic teams of graduate and senior undergraduate students with the opportunity to apply their knowledge to the analysis of big datasets. Each participating team will choose to analyze one of the two data sets described below. Each team is strongly encouraged to identify a faculty member to support the members as they develop their analytic approach and final presentation. Team members will work together to present a poster summarizing their methods and analysis results at the Annual Meeting.
Teams that select this case study will use the aggregated provincial-level hourly demand data for all sectors and annual demand data for each sector from The Canada Energy Regulator (CER) and the hourly air temperature and weather data from ETH Zurich and Imperial College London to develop statistical models to predict hourly electricity demand in the residential sector in Ontario.
Teams that select this case study will use podcasts from iTunes to develop statistical models to extract features from the provided podcasts and predict the number of reviews for the podcasts based on the extracted features.
One award will be presented for the best poster in each of the two case studies. The value of the award from SSC for each case study in the 2020 competition is 750 dollars with the expectation that this award is shared equally among the members of each winning team. An additional award with value of 750 dollars will be awarded by CER to the second best team working on the Case Study #1. The Committee reserves the right to decline to make an award for each case study if the number of entries is insufficient.
SSC also provides a total of $3,000 travel award for all participating teams excluding the winning teams. It is expected that this award will be shared equally among the participating teams excluding the winning teams for both of the case studies.
Depending on the number of participating teams in each of the two case studies, 1-2 teams for each of the case studies will be selected as Honorable Mention. There are no cash prize for honorable mentions.
All participating students will receive a certificate of participation.
TBA: Conference registration
Teams interested in participating in the competition MUST register the annual SSC meeting. We require each team should have at least one team member to register the meeting (using student registration rate). Please check the deadline for the conference registration at https://ssc.ca/en/meetings/annual/2020-annual-meeting
May 11, 2020: Competition registration
Teams interested in participating in the competition must register by this date by e-mailing the Chair of the Case Studies in Data Analysis Committee, Dr. Pingzhao Hu (Pingzhao.email@example.com). The registration information should include: Names and emails of team leader, team members and faculty mentor(s), university name, case study number, presentation title and indication of the team member(s) who registers the SSC conference.
May 18, 2020: Case Study 2 prediction result submission
Teams interested in participating in the Case Study 2 must submit their prediction result (an excel file with two columns: podcast ID, the number of predicted reviews. Please do not round the results into integer numbers) for each of the podcast in the unlabeled data set by this date to Dr. Kathryn Morrison (firstname.lastname@example.org). Please name the file as CaseStudyNumber_TeamLeaderFullName_UniversityName.xlsx. This prediction result will take 30% of your final poster judging score.
May 18, 2020: Abstract and group photo submission
Teams interested in participating in the competition must submit an abstract (maximum 500 words) with Sections of Introduction, Objective, Methods, Results and Conclusions by this date to the Chair of the Case Studies in Data Analysis Committee, Dr. Pingzhao Hu (Pingzhao.email@example.com). The abstract template can be downloaded here. A group photo for each team should be also submitted by this date. Please name the files as CaseStudyNumber_TeamLeaderFullName_UniversityName.
We require that the number of team members (either undergraduate students or graduate students) in a team should be not more than 4.
Each poster is recommended to contain the following information:
- Title of poster
- Names of team members and university affiliation(s)
- Results/Main findings (use figures, tables, and text)
- Conclusions (including strengths and limitations of your analysis)
You should acknowledge your team’s faculty mentor (if you have one) on your poster. The role of your faculty mentor is to provide advice and suggestions about your analysis, not to do the analysis for you.
The maximum size for your poster is 4 feet high by 3.5 feet wide (may change based on local organizers’ facilitation).
Consider the elements of good poster design as you prepare for the competition. Some useful resources are:
The Committee of the Award for Case Studies in Data Analysis will consider such attributes as result accuracy, innovation of the analysis methods, technical clarity, and cohesiveness of the analysis, interpretation and presentation of results in choosing winning teams.
Each poster will be evaluated based on the pre-defined criteria (see the Poster Judging Form here by two judging teams (2 judges per team). The judging teams will make consensus of the ranking of all participating teams for each of the two case studies, respectively.
Many thanks to members of the Case Studies in Data Analysis Committee for 2020 for their contributions: Dr. Kathryn Morrison, Precision Analytics Inc. and McGill University; Chel Hee Lee, Critical Care Medicine, Alberta Health Services & University of Calgary; Dr. Ehsan Karim, School of Population and Public Health, University of British Columbia; Dr. José Ribas Fernandes, Dr. Ryan Hum, Mr. Mantaj Hundal, Mr. Lukas Hansen, Mr. Michael Nadew, Mr. Matthew Hansen, Canada Energy Regulator.